Production of a Land Cover/Land Use (LC/LU) Map of Izmir Metropolitan City by Using High-Resolution Images

Abstract

The main objective of this study is to create a land cover/land use (LC/LU) map of a part in the Izmir metropolitan city in Turkey, based on an enhanced Urban Atlas nomenclature, using geographic object-based image analysis (GEOBIA) techniques. For this purpose, high-resolution SPOT 7 images from different seasons with a 1.5 m spatial resolution were used as the main earth observation (EO) data, and then rule-based object-oriented classification methods were applied to create the LC/LU map. Open-source geospatial data such as Open Street Map (OSM) and Wikimapia were also integrated into the classification to better identify some LC/LU classes such as Discontinuous Medium Density Urban Fabric, Fast Transit Roads and Associated Land, Airport as well as increase the total accuracy of the classification. In total, 23 LC/LU classes are determined in the study area with different accuracy values. An area-based accuracy assessment for random areas was performed to determine the classification accuracy of the obtained LC/LU map. The overall accuracy and kappa values were found as 88.84% and 0.882, respectively.